Three-dimensional video images are classified into two types: a view image and a depth map. The view image includes information, such as color information, of the image itself. The depth map includes depth information, such as contour information, of the view image. To reduce bandwidth occupied by video transmission during transmission of the three-dimensional video image, coding processing needs to be performed on the view image and the depth map so as to reduce a data size of the three-dimensional video image.
Simplified depth coding (SDC) is a method for coding a depth map. In a case in which an image block of the depth map on which the SDC needs to be performed is determined, a coder needs to select an intra-frame prediction mode and determine prediction data of the image block according to the intra-frame prediction mode. A prediction mode that can be selected by the coder may be any one of the following modes: a direct current mode, a planar (Planar) mode, and a depth modeling mode (Depth Modeling Mode, DMM), where the DMM includes an explicit wedgelet (Wedgelet) mode, an intra-predicted wedgelet partitioning mode, an inter-component prediction of wedgelet partitioning mode, and an inter-component prediction of contour partitioning mode. When determining the prediction data of the image block according to the intra-frame prediction mode, the coder needs to determine a size of the image block. When the SDC is performed on the image block, a size of a prediction block on which SDC is performed needs to be consistent with the size of the image block. However, in the prior art, a maximum prediction size (that is, a maximum size of the image block on which the SDC is performed) is limited, and the size of the image block may be greater than the maximum prediction size, for example, the size of the image block is 64×64, whereas the maximum prediction size is 32×32. In this case, when SDC is performed on the image block, an image block of a size greater than the maximum prediction size needs to be introduced into a coding and decoding system, which may increase design overheads of the coding and decoding system. In addition, the DMM does not support the image block of a size greater than the maximum prediction size. Therefore, in a case in which the size of the image block is greater than the maximum prediction size, the DMM cannot be selected to perform the SDC on the image block, thereby resulting in a decrease in coding efficiency.